Noninvasive Assessment of Heart Failure Using Features Derived from Cardiac Signals

Authors

Valentina Dargam, Florida International UniversityFollow

Document Type

Dissertation

Degree

Doctor of Philosophy (PhD)

Major/Program

Biomedical Engineering

First Advisor's Name

Joshua D. Hutcheson

First Advisor's Committee Title

Committee chair

Second Advisor's Name

Steven Bibevski

Second Advisor's Committee Title

Committee Member

Third Advisor's Name

Armando Barreto

Third Advisor's Committee Title

Committee Member

Fourth Advisor's Name

Zachary Danziger

Fourth Advisor's Committee Title

Committee Member

Fifth Advisor's Name

Robert Hacker

Fifth Advisor's Committee Title

Committee Member

Sixth Advisor's Name

Wei-Chiang Lin

Sixth Advisor's Committee Title

Committee Member

Keywords

engineering, biomedical engineering and bioengineering

Date of Defense

6-25-2024

Abstract

Recent advances in technology and scientific discovery have expanded the range of surgical interventions and pharmacotherapies for heart failure (HF). Despite these advances, HF outcomes remain poor. HF occurs when conditions damage or weaken the heart and is the end-stage of various cardiovascular diseases. Chronic kidney disease (CKD), which impairs renal function, frequently co-occurs with HF, and shared etiological factors worsen both conditions, leading to poorer prognoses. Early diagnosis and management are critical to preventing adverse cardiac remodeling and irreversible functional impairment in patients with CKD-induced HF. Diagnosing HF is challenging due to the lack of a single diagnostic test and the reliance on patient-reported symptoms, which are often underreported or absent even when HF is severe. Symptom-independent screening strategies are needed to address HF underdiagnosis.

Cardiac electrical and acoustic signals, measured via electrocardiograms and phonocardiograms, provide valuable information for detecting cardiac abnormalities. We hypothesize that differences in cardiac structure and function that lead to HF can be identified using these signals. The goals of this dissertation are to: 1) identify the relation between cardiac remodeling and function to acoustic and electrical signals, and 2) analyze changes in these signal characteristics and cardiac function during the progression of CKD-induced HF.

Aim 1 investigates whether aortic valve structural alterations induce changes in the second heart sound (S2) characteristics in a mouse model of CKD-induced valvular calcification. Using a machine-learning algorithm, we found that S2 characteristics can identify the presence of AoV calcification independent of hemodynamic changes. In aim 2, we determine if CKD-induced cardiac dysfunction can be identified and monitored using electrocardiogram signals. We identified sex-dependent changes in cardiac structure, function, and electrophysiology throughout the progression of CKD-induced HF. Aim 3 investigates the relation between S2 characteristics and markers of cardiac remodeling and right-sided heart dysfunction. We identified that S2 duration could predict elevated right-ventricular pressure.

This research aims to determine whether new markers derived from cardiac acoustic and electrical signals can be used to identify cardiac markers indicative of HF. This work could improve HF management by enabling early detection, preventing disease exacerbation, and identifying asymptomatic patients.

Identifier

FIDC011190

ORCID

https://orcid.org/0000-0002-6246-296X

Previously Published In

Dargam, V., Ng, H. H., Nasim, S., Chaparro, D., Irion, C. I., Seshadri, S. R., Barreto, A., Danziger, Z. C., Shehadeh, L. A., & Hutcheson, J. D. (2022). S2 Heart Sound Detects Aortic Valve Calcification Independent of Hemodynamic Changes in Mice. Frontiers in cardiovascular medicine, 9, 809301. https://doi.org/10.3389/fcvm.2022.809301

Recommended Citation

Dargam, Valentina, "Noninvasive Assessment of Heart Failure Using Features Derived from Cardiac Signals" (2024). FIU Electronic Theses and Dissertations. 5327.
https://digitalcommons.fiu.edu/etd/5327